View Full Version : Zigbee Transceivers
MichaelCollier
25-01-2006, 15:16
Hello,
I have found different zigbee transceivers on the internet and they vary in price. Does anyone know if the cheaper ones work with a picaxe?
I have included as much information about each transceiver in the links below. Thanks, Mike.....
--------------------------------
MC13191
Datasheet:
http://www.freescale.com/files/rf_if/doc/data_sheet/MC13191DS.pdf
Freescale Semiconductor:
http://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=MC13191&nodeId=0175200308816 6
Bought from UK distributor Digi-Key:
http://uk.digikey.com/scripts/dksearch/dksus.dll?KeywordSearch&site=uk&vendor=375&mpart=M C13191FC
£ (GBP) 3.18 each + £ (GBP) 12.00 delivery
----------------------------------------
CC2420
Datasheet:
http://www.chipcon.com/files/CC2420_Data_Sheet_1_3.pdf
Bought from Mouser:
http://www.mouser.com/index.cfm?handler=displayproduct&lstdispproductid= 706094
$ (US Dollar) 3.98 each + $ (US Dollar) 14.00
--------------------------------
Maxstream XBee Module
Picaxe £ (GBP) £14.00 + delivery.
---------------------------------
Stan. SWAN
25-01-2006, 15:51
Yikes - p&p ~3 times the ZigBee cost! Rev. Ed have settled on the 2.4GHz XBee & are shortly to release a complete 18X driven PCB module. I've been trialling a pair down here in sunny NZ, & found setup initially challenging. Xbees at least are akin to Hayes style modems & need Max232 driving chips for config.- good old 433MHz "plug & play" is a breeze in comparison ! I'd hence advise sticking with Rev.Ed's Xbee module & Picaxe wizard initially- prices are rumoured to be attractive!
These gave ~30m range indoors,with all the usual WiFi type 2.4GHz propagation hassles (metallic obstructions, people & vegetation especially). UNOBSTRUCTED ranges even with the 1mW version were amazing however(~2 km LOS with antenna =>www.picaxe.orcon.net.nz/zigscoop.jpg <A href='http://www.picaxe.orcon.net.nz/zigscoop.jpg ' Target=_Blank>External Web Link</a>
Both transceivers look like thay may be usable with a PICAXE. Their function seems to be to take a data payload and throw it in the air and to pull other data payloads back, with the need for additional processing of that payload.
I have no idea what effort decoding such payloads would have, how quickly it would need to be done, or what steps would need to taken to make them compatible and not interfere with other Zigbee transmissions. If it really is a case of just creating a payload and it turns up at the other end then it should be possible to use with a PICAXE.
What the XBee modules give is an integrated solution which deals with all the payload handling and allows a simple serial in and serial out connection, and it comes on a PCB which can be reasonably easily connected to a PICAXE. That's where the extra cost comes in, the others are just chips and all the support components and processing has to be added; XBee does that for you.
frankeltham
25-01-2006, 19:33
Aha, this is where the trancseiver talk has moved to.
Stan. Without using your seives what are the XBee ranges in the open? And have you tried Xbee Pro? (30m indoors up to a WHOPPING 2km outdoors? Is your tape-measure accurate? Or is that an I-can-just-about-hear-something 2km?)
Hippy. When you say payloads, is that your own jargon for 'packets'?
I was after easy-to-use transceivers with 600m plus range (open LOS). The CC2420 is completely OTT for what I want, which, in essence, is a sort of slow network. i.e. where each Transceiver would have its own ID and it has a listen before it transmits.
KIS. I see Xbee has IDs but not sure how much RxTxing is built in or how much the PICAXE would have to do. After all, you could do your own packetising and coding.
At the end of the day it ain't difficult as long you (or I) know precisely what each 'component' is doing. And I don't becuase the 'datasheet' link from the tech-supplies page doesn't work!!
Stan. SWAN
25-01-2006, 21:33
You maybe missed this, but 30m indoors thru' typical light timber framed NZ buildings while LOS outdoors bare ~ 200m. ANY 2.4GHz SIGNAL WILL BE HUGELY ATTENTUATED BY METAL OR VEGETATION. Even shieling with your body at 50m LOS blocked signals I found
Re "Payload"; yes "packet" is applicable. Having had a longer read through of each datasheets, although not to the depth that's probably necessary to confidently comment, the MC13191 looks like the easier of the two to use with a PICAXE.
MichaelCollier
26-01-2006, 10:07
If anyone needs the datasheet for the/an XBEE here is a link that works...
http://www.maxstream.net/products/xbee/product-manual_XBee_OEM_RF-Modules.pdf
frankeltham
26-01-2006, 13:37
Well done Michael.
Very interesting Mr Bond.. and it can duty cycle by itself.
Excellent.
Been doing a lot more reading on ZigBee and I think I've got to grips with it. It's very much like Ethernet but wire-free.
The MC13191 is effectively the network card's 'UART'; fill it up with bytes and it will send them out as a packet, when it sees some packet in the air it pulls it down and indicates it's got one. The MC13191 really doesn't care what the packet is or what it contains, only that it's validly framed and looks like a packet. It's like using a PICAXE and 433MHz to throw packets around.
Controlling the MC13191 is defined in another document ...
http://www.freescale.com/files/rf_if/doc/ref_manual/MC13191RM.pdf
At this level we don't really have "ZigBee", but we do have a mechanism to send a packet into the air and have it turn up elsewhere. So if that's all that's required that's all that's needed, but the configuration isn't compliant with ZigBee, can interfere with genuine ZigBee devices and be interfered with by them.
There's a layer above this raw packet transfer mechanism; the "ZigBee Protocol" or "ZigBee Stack". This defines what must appear where in the packets, where particular data must be placed so a ZigBee device can indicate who is sending the data, where it's destined for etc, etc, and to let the receivers know if it should take the incoming packet or discard it. Devices which comply with this protocol can all exist alongside each other without compromising the entire ZigBee network.
It's entirely possible to define one's own ( ie, simple ) protocol which uses the MC13191 to pass packets around which is incompatible with ZigBee and use that, but that obviously precludes interaction with compliant ZigBee devices and may give problems both ways if such devices are within range of each other. Just like connecting a PICAXE directly to an Ethernet LAN without complying to its rules, things are going to get messed up.
With 433MHz, everyone designs their own protocol and packet formats, and unless well designed, your neighbour's scheme may well interfere with your own data transfers and vice-versa. The ZigBee protocol attempts to overcome that by creating rules which mean neigbours don't have to resort to fist fights to win their space on the airwaves. Not complying with those rules is therefore not recommended, but could be done. There may however be local legislation which says it can't.
So, you have an MC13191, and then you need a controller which can take data from somewhere, setup packets correctly and untangle incoming packets to only pass on data which is directed to it. That controller either implements the ZigBee protocol or your own 'bodge-it' protocol. In most cases a separate controller would be used to do this ( effectively what the XBee gives ready made ), but it is possible that a PICAXE could use an MC13191 directly as if it were a 433MHz transceiver, ZigBee compliant or otherwise.
Whether a PICAXE and MC13191 solution would work I can't answer absolutely, but it is potentially possible. A lot depends upon how easy it is to configure the MC13191 to do its stuff, and the environment it's in. If there are ZigBee devices around throwing out their data, could the PICAXE sift through incoming packets quickly enough to find the ones directed at it ? Probably not, but if the airwaves were otherwise free I don't see a problem.
It's more than just connecting up a PICAXE and MC13191 though. There's hardware design ( the antenna circit alone ), initialisation and then protocol design, and understanding and implementing the ZigBee stack if seeking compliance.
It's an interesting project to attempt, but I doubt it's one for the faint hearted or anyone who isn't prepared to put a lot of effort into it. The extent of the effort involved is what will make most people look at XBee as being cost-effective. If someone did produce a "this will work" solution for a cheaper ZigBee chip, I'm sure it would be appreciated.
The most potentially useful application of direct MC13191 use I can see would be in using it as a transmission only device, which discarded all incoming packets and didn't handle error correction and re-transmission parts of the ZigBee protocol. Just like bit-banging a packet onto an Ethernet LAN, as long as what goes out is compliant with what it should be then it'll get to the other end if there's no collision or corruption. This is really souped-up 433MHz transmission. There's the same hardware design and chip control issues, but it would seem to be a lot more feasible to me.
Edited by - hippy on 1/26/2006 2:57:28 PM
MichaelCollier
26-01-2006, 15:58
Microchip have a free zigbee stack that they have developed, it is not entirely zigbee compliant though. Maybe of interest. Thanks for all the help so far.
PDF for it here...
http://ww1.microchip.com/downloads/en/appnotes/00965a.pdf
Okay, there's another aspect of ZigBee I hadn't realised ... ZigBee Profiles.
Once we have data going into and coming out of a ZigBee module, what does it mean ? It's a stream of raw bytes, and another layer up has to determine what the stream means and what the component parts of the data part of the transmitted packets are.
This is done by "ZigBee Profiles" which specify where and what data there is in a packet. This is what will allow commercial products which contain ZigBee devices to intereoperate, so a ZigBee Remote Lighting Controller can control a ZigBee Controlled Lighting Switch from another manufacturer and so on.
There's also a concept of "End Points" which also ties in with the Ethernet analogy I gave ( and also USB ), but this time we're up at the TCP/IP level. This allows a variety of input streams to connect to a variety of output streams and vice-versa, so any number of processes running in a product's controller can talk over ZigBee to a variety of clients or servers elsewhere without interfering with each other's data communications. It also allows our remote controller buttons to be directed to affect one or more switches in the lighting example.
There is also a special end point which can be interrogated to find out what a particular ZigBee unit supports, and all this is used in a complex ZigBee network to allow routing and pass-on of data when ZigBee devices are too far apart to communicate and pass through relay devices.
The entire ZigBee stack is therefore quite complex to implement, but less so than BlueTooth or Wi-Fi. A brief, but good, overview can be found here ...
http://wireless.industrial-networking.com/articles/articledisplay.asp?id=1008
Try this place for a transciever unit,
STAN, you might be interested in this one, the 433mhz transceiver for $30AUS, check the data sheet, two freqs, hi and low signal strength
Stan, is that you in the shorts??
Stan. SWAN
27-01-2006, 17:49
Shorts!- maybe shots (knowing my legs!)- & which place stocks these ?!! Oatley Electronics <A href='http://www.oatleyelectronics.com/pdf/rf-nrf401.pdf ' Target=_Blank>External Web Link</a> sell something akin of course. The "shorts' man in the Zigbee scoop pix is Andrew "Brightspark" - not me.
Edited by - Stan. Swan on 2/2/2006 5:09:47 AM
Stan. SWAN
01-02-2006, 19:03
Rev. Ed have now released their 2.4GHz<b> AXE210 </b> XBee ZigBee units @ GBP45 a pair,which includes an 18X. Considering the basic 1mW XBee itself is US$19 (~GBP12) this seems a good price just as a development kit alone. As well as instant swap out with the much more powerful (100mW) & sensitive XBee Pro this same kit supports a <b>LS-40EB </b> GPS unit (GBP 29) module!
See detailed .pdfs =>www.rev-ed.co.uk/docs/axe210.pdf <A href='http://www.rev-ed.co.uk/docs/axe210.pdf' Target=_Blank>External Web Link</a> and tutorial =>www.rev-ed.co.uk/docs/axe210_xbee.pdf <A href='http://www.rev-ed.co.uk/docs/axe210_xbee.pdf' Target=_Blank>External Web Link</a> , with the units themselves sold as <b>AXE210 </b> , a.k.a a XBE010 "Starter Pack", via Tech-Supplies =>www.techsupplies.co.uk <A href='http://www.techsupplies.co.uk' Target=_Blank>External Web Link</a>
I've rustled up assorted cardboard & foil etc 2.4GHz antenna to suit, & found a 4-6dBi gain "bendy ruler" parabola most effective for the effort. See =>www.picaxe.orcon.net.nz/bendy.jpg <A href='http://www.picaxe.orcon.net.nz/bendy.jpg ' Target=_Blank>External Web Link</a>.
Naturally (!)larger cookware WiFi style parabola were also trialled, managing ~2km LOS across water =>www.picaxe.orcon.net.nz/zigscoop.jpg <A href='http://www.picaxe.orcon.net.nz/zigscoop.jpg ' Target=_Blank>External Web Link</a> - quite amazing for 1mW ! Stan
Edited by - Stan. Swan on 2/1/2006 6:19:41 PM
frankeltham
01-02-2006, 20:07
Do the shorts need to be of parabolic design too?
MichaelCollier
17-02-2006, 14:14
Ahh, I hadn't realised this thread was still active..
Thanks for the update Stan, I will go ahead and get AXE210 as soon as tech-supplies have upgraded their server.
I am still aiming at getting zigbee implemented using the lowest level components (and price), but for now I think I will need to get one or more development boards. The AXE210 seems the best value for money dev board I have seen so far.
I am hoping that the AXE210 will get me closer to the zigbee "architecture", I am also looking at microchips zigbee offerings, they have a zigbee dev board but charge over £100 for it, plus you need further hardware to program their board. A C-compiler is also required (500 dollars or free eval version) if you want to look at or modify their free zigbee stack - which I do intend to do.
If anyone has any details of using zigbee transceivers with the "minimum" of components I would be interested.
Cheers,
Mike
Stan. SWAN
17-02-2006, 20:24
The March & April Australian "Silicon Chip" mags should feature Picaxe/Zigbee config. articles by Rev.Ed's Clive,with antenna input by myself.
I've yet to come across a better starter or more cost effective ZigBee developers kit than Rev.Ed's AXE210, although the holy grail of "ZigBee light switches for peanuts" seems a while down the track yet. Stan
Powered by vBulletin® Version 4.1.10 Copyright © 2012 vBulletin Solutions, Inc. All rights reserved.